620 Thorax 1990;45:620-622 at altitude Thorax: first published as 10.1136/thx.45.8.620 on 1 August 1990. Downloaded from P S Thomas, R M Harding, J S Milledge

Abstract Methods The mini Wright peak flow meter is a Six healthy men sat in a decompression cham- useful, portable instrument for field ber at the RAF Institute of Aviation studies but being sensitive to air density Medicine, Farnborough. They breathed 40% will under-read at altitude. True peak oxygen when barometric pressure was expiratory flow will increase at altitude, 520 mmHg or less. Oxygen masks were however, because of the decreased air removed two minutes before they performed a density, given that dynamic resistance is respiratory test. The best of three PEF read- unchanged. The effect of simulated ings was recorded, a mini Wright peak flow altitude on peak expiratory flow (PEF) meter being used (Airmed, Clement Clarke was determined in six subjects with both International). Three forced the mini Wright meter and a volumetric (FVC) manoeuvres were recorded with a roll- spirometer (which is unaffected by air ing seal spirometer (Spiroflow, P K Morgan, density). With increasing altitude PEF as Chatham). The electronic output of the latter measured by the spirometer increased was interfaced with an Amstrad 1640 PC linearly with decreasing pressure, so that microcomputer to derive flow-volume loops at a barometric pressure of 380 mm JIg* for expiration and inspiration. These (half an atmosphere, corresponding to measurements were made at sea level at the an altitude of 5455 m) there was a 20% beginning and end of the experiment and at increase over sea level values. The mini three levels of decompression. The chamber Wright flow meter gave readings 6% was decompressed from sea level below sea level values for this altitude- (760 mm Hg) to the equivalent of 5455 m that is, under-reading by 26%. Meas- (380 mm Hg) over six minutes. Thereafter urements of PEF made at altitude with pressures equivalent to altitudes of 4242 m mini meter cor- mm m mm were the Wright should be (447 Hg) and 3030 (520 Hg) http://thorax.bmj.com/ rected by adding 6'6% per 100 mm Hg attained at 10 minute intervals before the drop in barometric pressure. return to sea level pressure. The total experimental time was about 90 minutes. The relation of PEF to barometric pressure The mini Wright peak flow meter' is a small, was modelled for each subject with the GLIM cheap, easily portable instrument weighing 3-77 software package,7 pressure being used as 75 g. It therefore lends itself to field the covariate to test the linearity of the measurements of peak expiratory flow (PEF),' relationship. on September 28, 2021 by guest. Protected copyright. a reproducible index of ventilatory function. the standard meter Using Wright peak flow % sea level Singh et al found a reduction in PEF in 23 of PEF 24 subjects who developed symptoms of 120- mountain sickness.3 Stockley and Green also showed a fall in PEF at altitude that was more pronounced in those who developed acute Spirometer mountain sickness.4 With the decrease in air density at altitude, 110 - however, the peak flow meter will under- read.56 The same decrease in air density, however, must result in an increase in the true PEF if dynamic airways resistance remains unchanged. When the peak flow meter is used Clinical Research 100 - Centre, Northwick at altitude these two effects would tend to Park Harrow cancel each other out. Hospital, PF \ P S Thomas To see if this is the case and to document Wright's J S meterr Milledge the change with altitude in PEF as measured RAF Institute of the we carried out a at Aviation Medicine, by spirometer, study 90- Farnborough simulated altitude, using the peak flow meter I 1 R M Harding and a spirometer to measure PEF and deter- 760 700 600 500 400 300 Address for reprint requests: mine the maximum expiratory flow-volume Barometric pressure (mm Hg) Dr J S Milledge, Clinical curve. Research Centre, Northwick Figure 1 Peak expiratoryflow (PEF) as percentage of Park Hospital, Harrow sea level valuefor six subjects measured by spirometer HA1 2UJ (0) and by the mini Wright peakflow (PF) meter (0) Accepted 24 May 1990 *1 mm Hg t 0-133 kPa. at simulated altitude. Bar lines indicate 1 SEM. Peak expiratoryflow at altitude 621

Peak expiratoryflow measurements (I/s) obtainedfrom a spirometer and a mini Wright with decreasing pressure to show a 20% peakflow meter at sea level (SL) and three simulated altitudes increase at 380 mm Hg (5455 m). There was a small, steady fall in the PEF measured by the Subject No 760 (SLI*) 380 (5455) 447 (4242) 520 (3030) 760 (SL2*) Mean SL peak flow meter with increasing altitude. Both

SPIROMETER these changes of PEF with pressure were Thorax: first published as 10.1136/thx.45.8.620 on 1 August 1990. Downloaded from 1 8-46 10-57 9-71 9-54 9-27 8-86 significant (p < 0 001). 2 8-21 11-71 10-23 10-77 8-78 8-50 Modelling the results from both instruments 3 8-74 9-23 11-15 10-23 9-41 9-07 4 7-51 10-78 9-63 10-54 10-28 8-89 showed that the relationship of PEF with 5 10-32 13-13 11-86 11-28 10-04 10-18 barometric pressure was linear, the slope being 6 9-96 10 90 11-67 10-60 10 00 9-98 negative for the spirometer and positive for the Mean 8-87 11-05 10-71 10-50 9-63 9-25 peak flow meter. The mean and SD of the SD 1-07 1-29 0-98 0-58 0 57 0-67 slopes for change in PEF (l/s) per mm Hg MINI WRIGHT PEAK FLOW METER pressure were: 1 8-50 8-33 8-17 8-33 8-50 8-50 Spirometer -0-0048 (0 0008) 2 9 00 9 00 8-83 8-67 9-17 9-08 3 10-00 9-17 9-50 9-67 10 00 10-00 Peak flow meter + 0-0016 (0 0004). 4 10-50 9 33 10-17 11-17 11-67 11-08 The relation with pressure for individual 5 10-17 9-50 9 50 9-67 10 00 10-08 subjects was linear (rather than, for instance, 6 9-83 9 33 9-17 9-50 9-67 9-75 Mean 9-66 9 11 9-22 9-50 9-83 9 75 quadratic). SD 0-76 0-40 0-67 1 00 1-05 0-88 The mean maximum expiratory flow- volume curves for sea level and 5455 m *1 and 2 indicate before and after decompression. (380 mm Hg) show that the increased flow at altitude continues down to 25% of vital capacity (fig 2). Maximum inspiratory flows Results were also increased at altitude. There was no No subject suffered side effects or noticed significant change in FEV1 or FVC. symptoms of altitude sickness. The results are summarised in the table and figure 1. PEF at sea level pressure measured Discussion with the peak flow meter did not differ sig- The spirometer measured PEF rises progres- nificantly from PEF measured with the sively with decreasing barometric pressure spirometer, nor was there any significant dif- when there is no change in dynamic airways ference between sea level values at the begin- resistance. (Being a volumetric instrument, the ning and end ofthe study (table). Subject 4 had spirometer is unaffected by changes in air no previous experience of FVC manoeuvres density.) At a pressure of half an atmosphere and seemed to show a learning effect through (380 mm Hg) PEF has increased by 20%. The the study. His first PEF (spirometer and meter) flow at volumes down to 25% FVC was was probably below his true value. also found to be increased (fig 2). This is in line http://thorax.bmj.com/ The PEF measured by spirometer increased with the results ofstudies using helium-oxygen

Figure 2 Mean 160 - maximum expiratoryflow- Flow volume loops ofsix subjects (I/s) at sea level and at simulated altitude 14 140 - on September 28, 2021 by guest. Protected copyright. (380 mm Hg pressure). * Sch ilder 1963 Bar lines indicate I SEM. o Present study 120 - LL-

100 - 0, a1) 80 -

60 -

40-

. 0 1 2 3 4 %, Gas density

0 50 100 Figure 3 Effect ofchange of inspired gas density on peak expiratoryflow (PEF), air at sea level being taken Volume % FVC as unity: datafrom Schilder et aP and the present study. 622 Thomas, Harding, Milledge

Altitude (metres) flow meter that we found. The study by Singh et al does show that as acute mountain sickness O 1000 2000 3000 4000 5000 60007000 8000 I subsides PEF improves, suggesting that inter- 36 - I I I I I I I stitial pulmonary oedema was also contributing to the reduced PEF. Thorax: first published as 10.1136/thx.45.8.620 on 1 August 1990. Downloaded from The effect of gas density on PEF can be studied by inhalation ofgas mixtures ofvarying 32- density at sea level. Data derived from such a OTrue PEF study9 are plotted as percentage PEF change against gas density in figure 3. The relationship 28- *Wright mini PF meter is curvilinear. The values obtained in the present study (also shown in fig 3) fall on this line, suggesting that changes in PEF at altitude are accounted for by changes in air density. 24- Our results for change in spirometer aL) cn measured PEF with altitude and the correction to be applied to PEF measured with the peak Cu c 20- flow meter to obtain a "true" PEF for any given L* altitude or barometric pressure are shown in 00 a figure 4. This correction is in agreement with 0 the findings of Forster and Parker,5 who 0 calibrated the mini Wright peak flow meter L.. 16 - L. with a mechanical device at sea level and at C-,0 4200 m. They found an under-reading of20%, similar to the correction for this altitude given 12 - in figure 4 (21%). The relationship is linear with respect to pressure and therefore curvi- linear with respect to altitude. The change in meter measured PEF is about 50%/100 mm Hg 8- reduction in pressure. The correction to be applied to the peak flow meter is 6 6%/ 100 mm Hg pressure drop. By extrapolation it 4- will be seen that at the summit of Mount Everest (253 mm Hg) the spirometer measured PEF would be increased by 27% over sea level values (ifthere is no change in dynamic airways 0- resistance) and here any reading from the mini http://thorax.bmj.com/ Wright peak flow meter would have to be 800 700 600 500 400 300 200 increased by 33%. pressure Barometric (mml'Hg) We wish to thank the Commandant and staff of the Institute of Figure 4 Effect of reduced barometric pressure or altitude on spirometer measured peak Aviation Medicine at RAF Farnborough for their assistance in expiratoryflow (PEF), as percentage changefrom sea level value, and the correction to carrying out this study and for the use of their facilities, especially those staff members who acted as subjects and those be applied to PEF measured by a mini Wright peakflow meter to obtain true PEF (0). Note that the altitude scale is non-linear. who operated the chamber. on September 28, 2021 by guest. Protected copyright.

1 Wright BM. A miniature Wright peak-flow meter. Br Med J 1978;ii:1627-8. mixtures to change gas density, where the 2 Singh I, Khanna PK, Srivastava MC, Lal M, Roy SB, effect was seen down to non- Subramanyam CSV. Acute mountain sickness. N Engl J about 10% in Med 1969;280:175-84. smokers and 20% in smokers.8 3 Wright BM, McKerrow CB. Maximum forced expiratory flow rate as a measure of ventilatory capacity. Br Med J The mini Wright peak flow meter, which is 1959;ii:1041-7. sensitive to air density, under-reads by 26% at 4 Stockley RA, Green ID. Birmingham Medical Research half atmospheric pressure (380 mm Hg, Expeditionary Society 1977 expedition: cardiopulmonary function before, during and after a twenty-one day 5455 m) and gives readings for PEF 6% less Himalayan trek. Postgrad Med J 1979;55:496-501. 5 Forster P, Parker RW. Peak expiratory flow rate at high than at sea level. This latter result is similar to altitude. Lancet 1983;ii:100. the findings of Forster and Parker,5 who, using 6 Massen H, Boissinot E, Moiline J. Mesure du debit de pointe a mini flow found a reduc- et altitude. Cahiers d'Anesthesiologie 1986;34:341-3. Wright peak meter, 7 Payne CD, ed. The GLIM system release 3 77 manual. tion in PEF of6-8% in subjects taken rapidly to Oxford: Numerical Algorithms Group, 1986. 8 Hutcheon M, Griffin P, Levison H, Zamel N. A new test in 4200 m. detection of mild abnormalities of lung mechanics. Am The results of earlier studies showing a Rev Respir Dis 1974;110:458-65. reduction in PEF at altitude may be 9 Schilder D, Roberts A, Fry DL. Effect of gas density and viscosity on the maximal expiratory flow-volume relation- confounded by the under-reading of the peak ship. J Clin Invest 1963;42:1705-13.